Having worked in some DRDO labs before quitting in disgust, here are my thoughts
a)Promise the moon, but can't deliver a mustard seed. For example promise an air to air missile that is shorter, lighter, longer range more accurate and carries a heavier payload than any other missile. They just pick up Janes and make a Frankenstein missile by combining the features of 10 missiles. It's like asking for a wife who is tall&short, fair&dark, fat&thin... Contradictory requirements
b)It takes years to start. For example there is this program called RTA(regional transport aircraft) by NAL. I know it's not DRDO, but it's similar crap(unclassified though). They have spent the last 4 years deciding whether to make it a 60,70 or 100 seater. Israeli aircraft industries where consulted as a collaborator. They quit in disgust and said they could have made all 3 variants in 4 years.
c)Don't bug fix in the start. LCA has bad Aerodynamics. No matter how good the control system, it's just putting lipstick on a pig.
d)Pennywise and pound foolish (You won't trust scientists to drive to a meeting, afraid that they will overcharge you for fuel. Instead you will hire tourist taxis to ferry scientists around which will invariably delay meetings by a couple of hours-Time is MONEY)
e)Lack of internet. These labs are not fountains of knowledge that need to be guarded from the world. They are more like oceans of ignorance. Cut them off from the net and you are then a frog in a well.
f)A handful of smart scientists can do more than a 1000 idiots. A thousand monkeys on a typewriter can't churn out Shakespeare. Idiots slow everyone down. A few gems in an ocean of refuse will eventually sink in the muck.
g)Get the best scientists from Sukhoi, Mig etc to train your people. When you need outside help ask for it
i)No independent reviews and an old boys club mentality of the reviewers.